We all know someone who has fed their Facebook page with pictures of delicious and photogenic meals. Maybe you’ve done it yourself. In our digital age, more and more people are sharing pictures of their food, humble-bragging about their tasty adventures. Some public health researchers now have invented a tool to put this social media content to good use.

A team of scientists at the University of Vermont have developed a device called the Lexicocalorimeter, which gathers and analyzes tens of millions of tweets for words dealing with food and words dealing with activities such as “running, “watching TV” or “skiing.”

On a macro scale, the food words and activity words get scored based on calorie content data and caloric burn rates, respectively. Public health officials can use the data to get an idea of the overall health of a specific region or state based on tweets from those areas. The authors note that this calorie-to-activity ratio is not absolute, but it can be an important comparison tool.

Data from the Lexicocalorimeter is publicly available online and gives a snapshot of each state’s caloric input and output. The data shows that Colorado has the best balance between caloric input and output, with the words “noodles” and “running” dominating tweets. Mississippi ranks last with a high portion of posts containing “cake” and “eating.”

The team hopes this tool will help health care providers make data-driven recommendations for public policy through real-time assessments at the population level. The researchers say other similar tools in the works include the Insomniameter to measure sleep deprivation and, possibly, a Hangoverometer.